Comparison of two-dimensional gel electrophoresis patterns and MALDI-TOF MS analysis of therapeutic recombinant monoclonal antibodies trastuzumab and rituximab

Journal of Pharmaceutical and Biomedical Analysis 56 (2011) 684–691

Contents lists available at ScienceDirect

Journal of Pharmaceutical and Biomedical Analysis journal homepage: www.elsevier.com/locate/jpba

Comparison of two-dimensional gel electrophoresis patterns and MALDI-TOF MS analysis of therapeutic recombinant monoclonal antibodies trastuzumab and rituximab Dashnor Nebija, Hannelore Kopelent-Frank, Ernst Urban, Christian R. Noe, Bodo Lachmann ∗ Department of Medicinal Chemistry, Faculty of Life Sciences, University of Vienna, Vienna, Austria

a r t i c l e

i n f o

Article history: Received 1 April 2011 Received in revised form 25 May 2011 Accepted 6 July 2011 Available online 18 July 2011 Keywords: Trastuzumab Rituximab 2-DE MALDI-TOF MS PMF

a b s t r a c t The principal objective of this study was the evaluation of two-dimensional gel electrophoresis (2-DE) in combination with MALDI-TOF MS, after tryptic digest with regard to suitability for qualitative characterization and identification of therapeutic recombinant monoclonal antibodies trastuzumab and rituximab. Moreover, the impact of post-translational modifications of these glycoproteins on the electrophoresis behavior has been evaluated. 1-D SDS-PAGE, in reducing and non-reducing conditions, and 2-DE were used for the assessment of Mr and the monitorization of deglycosylation efficiency. In addition, 2-DE was used for the determination of pIs. 2-DE gels revealed characteristic glycoprotein migration behavior, highly complex spot pattern, typical for recombinant monoclonal antibodies. N-linked oligosaccharides were released with PNGase F; enzymatic desialination was studied with sialidase and carboxypeptidase B was used for the study of lysine truncation. Peptide spots resolved in 2-DE gels were in gel tryptically digested, resulting peptides were subjected to MALDI-TOF MS analysis and peptide mass fingerprinting (PMF) has been used for the identity confirmation of both monoclonal antibodies. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Trastuzumab, [FDA, USAN, INN], Herceptin® [TN], Syn: huMab 4D5-8, is a humanized recombinant monoclonal antibody (rmAb), produced in Chinese Hamster Ovary (CHO) cells. It selectively targets the extracellular domain of the human epidermal growth factor receptor 2 protein (HER-2). The antibody is an IgG1 that contains human framework regions with the complementaritydetermining regions (CDR) of a murine anti-p185HER2 antibody that binds to HER-2 [1,2]; see Fig. 1. Rituximab (FDA, USAN, INN), Rituxan® , MabThera® (TNs) is a genetically engineered chimeric rmAb, produced in CHO cells. This IgG1-␬ antibody contains murine light and heavy chain variable regions, and human gamma 1 heavy chain and kappa light chain constant regions; see Fig. 2 Both rmAbs are composed of four polypeptide chains linked via disulfide bonds: two light and two heavy chains. Trastuzumab contains 1328 amino acids, its theoretical molecular weight (Mr ), calculated from the amino acid sequence of the

∗ Corresponding author at: Department of Medicinal Chemistry, Faculty of Life Sciences, University of Vienna, Althanstrasse, 14, 1090 Vienna Austria. Tel.: +43 01 4277 55065; fax: +43 01 4277 9551. E-mail address: [email protected] (B. Lachmann). 0731-7085/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jpba.2011.07.006

reduced non-glycosylated form is 145.45 kDa, however, the apparent Mr is higher due to the presence of N-linked oligosaccharides [7]. Each heavy chain contains 450 amino acids and each light chain 214 amino acids, with their respective theoretical Mr , 49.284 kDa and 23.443 kDa and respective theoretical pI values, 8.49 and 7.76, calculated using PeptideMass program [8]. Trastuzumab contains only one N-glycosylation site at the conserved heavy chain asparagine 300 [3]. Rituximab contains 1328 amino acids, the theoretical Mr of reduced non-glycosylated form is 144.544 kDa and the apparent Mr is higher due to the heavy chain glycosylation at asparagine N-301 [9]. Each heavy chain contains 451 amino acids, and each light chain contains 213 amino acids, with their respective theoretical Mr 49.214 kDa and 23.057 kDa and respective theoretical pI values 8.67 and 8.26. Glycosylation of rmAbs has no effect on antigen binding. However, it is required for the Fc mediated effector functions such as antibody-dependent cellmediated cytotoxicity and complement dependent cytotoxicity (CDC) [10]. Moreover, rapid elimination of IgG-antigen complexes from circulation is strongly influenced by IgG glycosylation [11]. Trastuzumab and rituximab exhibit glycane profiles basically consisting of asialo, neutral, fucosylated, complex biantennary oligosaccharide type, as determined by CE-LIF of PNGase released glycans, after derivatization with 3-aminobenzoic acid,

D. Nebija et al. / Journal of Pharmaceutical and Biomedical Analysis 56 (2011) 684–691

685

DIQMTQSPSS LSASVGDRVT ITCRASQDVN TAVAWYQQKP GKAPKLLIYS

Light chain:

ASFLYSGVPS RFSGSRSGTD FTLTISSLQP EDFATYYCQQ HYTTPPTFGQ GTKVEIKRTV AAPSVFIFPP SDEQLKSGTA

SVVCLLNNFY

LSKADYEKHK

VYACEVTHQG

LSSPVTKSFN

LVQPGGSLRL

SCAASGFNIK

DTYIHWVRQA

PGKGLEWVAR

IYPTNGYTRY

ADSVKGRFTI

SADTSKNTAY

LQMNSLRAED

TAVYYCSRWG

GDGFYAMDYW

GQGTLVTVSS

ASTKGPSVFP

LAPSSKSTSG

GTAALGCLVK

DYFPEPVTVS

WNSGALTSGV

HTFPAVLQSS

GLYSLSSVVT

VPSSSLGTQT

YICNVNHKPS

NTKVDKKVEP

KSCDKTHTCP

PCPAPELLGG

PSVFLFPPKP

KDTLMISRTP

EVTCVVVDVS

HEDPEVKFNW

YVDGVEVHNA

KTKPREEQYN

STYRVVSVLT

VLHQDWLNGK

EYKCKVSNKA

LPAPIEKTIS

KAKGQPREPQ

VYTLPPSREE

MTKNQVSLTC

LVKGFYPSDI

AVEWESNGQP

ENNYKTTPPV

LDSDGSFFLY

SKLTVDKSRW

QQGNVFSCSV

PREAKVQWKV

DNALQSGNSQ RGEC;

ESVTEQDSKD

STYSLSSTLT

and heavy chain:

EVQLVESGGG

MHEALHNHYT QKSLSLSPGK *

Complementarity determining regions highlighted and the heavy chain N300

glycosylation site indicated in bold type. Fig. 1. Amino acid sequence and CDRs of trastuzumab [3,4].

2-aminobenzoic acid, and 1-aminopyrene-3,6,8-trisulfonate, respectively. With these techniques, only minor amounts of terminal sialinated oligosaccharides (1–5% of total glycan amount) have been detected [12–15]. Furthermore, CE-ESI-MS and LC-ESI-MS methods were used for the glycosylation studies of these rmAbs [9,16,17].

Charged variants of rmAbs can be studied using different methods, such as high-performance ion exchange chromatography (HPIEC), isoelectric focusing (IEF), and capillary zone electrophoresis (CZE) [18]. Charged variants which can occur at the four N-termini and the two heavy chain C-termini of rituximab were investigated with HPIEC, after papain digestion [19]. For the

Light chain: QIVLSQSPAI LSASPGEKVT MTCRASSSVS YIHWFQQKPG SSPKPWIYAT SNLASGVPVR FSGSGSGTSY SLTISRVEAE DAATYYCQQW TSNPPTFGGG TKLEIKRTVA APSVFIFPPS DEQLKSGTAS VVCLLNNFYP REAKVQWKVD NALQSGNSQE SVTEQDSKDS TYSLSSTLTL SKADYEKHKV YACEVTHQGL SSPVTKSFNR GEC;

and heavy chain: QVQLQQPGAE LVKPGASVKM

SCKASGYTFT

SYNMHWVKQT

PGRGLEWIG A

IYPGNGDTSY

NQKFKGKATL

MQLSSLTSED

SAVYYCAR ST

YYGGDWYFNV

WGAGTTVTVS

AASTKGPSVF PLAPSSKSTS

GGTAALGCLV

KDYFPEPVTV SWNSGALTSG

TYICNVNHKP SNTKVDKKAE

PKSCDKTHTC

PEVTCVVVDV

WYVDGVEVHN

SHEDPEVKFN

KEYKCKVSNK ALPAPIEKTI SKAKGQPREP IAVEWESNGQ

PENNYKTTPP VLDSDGSFF

VHTFPA

VLQS

TADKSSSTAY

SGLYSLSSVV TVPSSSLGTQ

PPCPAPELLG GPSVFLFPPK PKDTLMISRT AKTKPREEQY QVYTLPPSRD L

NSTYRVVSVL TVLHQDWLNG

ELTKNQVSLT CLVKGFYPSD

YSKLTVDKSR WQQGNVFSCS

VMHEALHNHY

TQKSLSLSPG K *

Murine

variable

domains

highlighted,

complementarity

determining

regions

underlined and italicized and the heavy chain N 301 glycosylation site indicated in bold type Fig. 2. Amino acid sequences of rituximab [5,6].

686

D. Nebija et al. / Journal of Pharmaceutical and Biomedical Analysis 56 (2011) 684–691

Fig. 3. 1-D SDS-PAGE analysis of trastuzumab and rituximab. (A) trastuzumab reducing conditions. Lanes 2–6, trastuzumab (1.3–20.5 ␮g), lanes 1, 7 MW standards; (B) rituximab, reducing conditions. Lanes 4–8, rituximab (0.1–5 ␮g), lanes 1, 10 MW standards; (C) trastuzumab, non-reducing conditions. Lanes 2–6 trastuzumab (1.3–20.5 ␮g), lanes 1, 7 MW standards; (D) rituximab: lanes 1, 9 MW standards; lanes 3, 5, reducing conditions (15 and 1.5 ␮g, respectively), lanes 7, 10, non-reducing conditions (15 and 1.5 ␮g, respectively); (E) deglycosylation pattern: abatacept (lanes 2, 3 and 4), trastuzumab (lanes 5, 6 and 7) and rituximab (lanes 8, 9 and 10), 2 ␮g each. Lane 1, MW standards, lanes 5, 8, glycosylated rituximab and trastuzumab, lanes 6, 7, 9 and 10, deglysolylated trastuzumab and rituximab (slightly different conditions); silver stained acrylamide gels: 12.5% T; 2.6% C.

determination of trastuzumab and rituximab serum levels, flow cytometry and ELISA have been described [20–22]. CDC assay was used for the determination of the rituximab potency [23], whereas potency of trastuzumab was determined by antiproliferation activity test [24]. High-performance liquid chromatography (HPLC) coupled with ESI-MS was used for the quantitation of intact trastuzumab [25] and for quantitative analysis of trastuzumab capillary isoelectric focusing (cIEF) and SDS-capillary gel electrophoresis (SDS-cGE) have been evaluated, demonstrating adequate sensitivity, precision and linearity for the use in a quality control environment [26]. Cation-exchange chromatography, a validated method used for the release of trastuzumab production lots, was used for the study of variants arising from the deamidation of asparagine and aspartate isomerization [3] and terminal lysine processing was studied using RP HPLC. In the latter case, cyanogen bromide cleavage following methionin 431, released C-terminal peptides that were studied with RP-HPLC [27]. Methionin oxidation in rmAb trastuzumab was studied by papain digestion of Fab and Fc fragments and by the detection of oxidized Fc fragments using hydrophobic interaction chromatography [28]. Peptide mapping was used to assess genetic heterogeneity of recombinant cell lines [29]. Furthermore, it has been widely recognized that 2-DE is a powerful method for the analysis of complex protein mixtures, able to be complemented with downstream MALDI-TOF MS analysis for identification purposes [30]. Additionally, the approval of 2-DE for

the routine quality assurance of recombinant drugs has been proposed [31]. In this study we have examined the suitability of 2-DE and subsequent PMF analysis for the study of identity, purity and post-translational modifications of therapeutically relevant rmAbs trastuzumab and rituximab. 2. Materials and methods Isoelectric focusing. 1 ml aliquot rehydration solution (urea 8 M, thiourea 2 M, CHAPS 4%, Triton-X100 0.5%, bromophenol blue 0.005%), 5 mg/ml DTT and 5 ␮l/ml IPG buffer of selected pH interval (GE Healthcare Bio-Sciences AB Uppsala, Sweden) were mixed prior to use. In 340 ␮l of this solution, 10 ␮l (1–5 ␮g/␮l) of sample solution were mixed and shortly vortexed. The mixture was allowed to stay 1 h at room temperature followed by centrifugation (15,000 g, 5 min). Strip Holders were put onto the cooling plate/electrode contact area of the IPGphor strip holder platform (IPGPhorTM IEF system, GE Healthcare, Biosciences AB, Uppsala, Sweden). Prepared samples were applied onto commercially available Immobiline DryStrip gels (6–9 L and 3–11 NL, length 18 cm ±2 mm, GE Healthcare Bio-Sciences AB Uppsala, Sweden) by ingel rehydration [32]. The temperature was set to 20 ◦ C and the current limited to 50 ␮A/I PG strip. Focusing was run to a total 72 kVh. Second dimension, SDS-PAGE. The second-dimension, SDS-PAGE, was run on a vertical system (PROTEAN® II Xi Cell for vertical

D. Nebija et al. / Journal of Pharmaceutical and Biomedical Analysis 56 (2011) 684–691

687

Fig. 4. 2-DE analysis of trastuzumab and rituximab. (A) IPG 3–11 NL, non-deglycosylated, 10 ␮g trastuzumab; (B) IPG 6–9 L, non-deglycosylated, 10 ␮g trastuzumab; (C) IPG 3–11 NL, deglycosylated, 50 ␮g trastuzumab; (D) IPG 3–11 NL, non-deglycosylated, 10 ␮g rituximab; (E) IPG 6–9 L, non-deglycosylated, 10 ␮g rituximab (F), IPG 3–11 NL, deglycosylated, 50 ␮g rituximab. Silver stained acrylamide gels: 12.5% T; 2.6% C.

electrophoresis, 20 cm × 20 cm, Bio-Rad Laboratories, USA). Focused IPG strips were equilibrated twice, each time for 15 min, by gently shaking. The first equilibration step was performed in 5 ml equilibration buffer (Tris 0.05 M, urea 6 M, glycerol 30%, SDS 2%, bromphenol blue 0.005%) containing 1% DTT, followed by the

second equilibration in 5 ml equilibration buffer containing 2.5% iodoacetamide [33]. Equilibrated IPG gel strips were placed on top of the vertical SDS gel. Molecular weight standards (Precision Plus Unstained ProteinTM Standards, Bio-Rad Laboratories, USA) were placed at

Fig. 5. MALDI-TOF MS spectra (peptide mass fingerprints) of peptides generated from deglycosylated, 2-DE separated and in situ tryptically digested trastuzumab light (up) and heavy chain (down).

688

D. Nebija et al. / Journal of Pharmaceutical and Biomedical Analysis 56 (2011) 684–691

Fig. 6. MALDI-TOF MS spectra (peptide mass fingerprints) of peptides generated from deglycosylated, 2-DE separated and in situ tryptically digested rituximab light (up) and heavy chain (down).

the anodic end of the SDS-PAGE gel. Second dimension electrophoresis run (lab made gels 12.5% T, homogenous, 2.6% C; 200 mm × 200 mm × 1 mm) was performed under constant current of 45 mA/gel. One-dimensional, SDS-PAGE was performed according to Laemmli [34], on the mini-gel system Mini-PROTEAN® 3 Cell, 8 cm × 7.3 cm, (Bio-Rad Laboratories, USA). Silver staining was performed according to Blum et al. [35]. Coomassie staining (Coomassie Brilliant Blue G 250, Serva Electrophoresis GmbH, Heidelberg Germany) was performed according to the manufacturer’s instructions. All chemicals used were reagent and HPLC grade. Gels were scanned (Biorad Densitometer GS 710) and analyzed with PDQuest 6.2.1. (Bio-Rad Laboratories, USA). In-gel digestion. For silver stained gels, in situ reduction, alkylation and tryptic digestion of proteins resolved from 2-DE gels, with prior destaining, was performed according to methods described elsewhere [36,37]. Sample redissolution and MALDI-TOF sample presentation. Dry pellets of digestion extracts were resuspended in 5 ␮l, 0.1% TFA. Peptide mixtures were analyzed using a saturated matrix solution of ␣-cyano-4-hydroxycinnamic acid (Sigma–Aldrich Chemie GmbH, Steinheim, Germany) in 70% ACN: 30% of 0.1% TFA (v/v). All mass spectra were acquired on Compaq SEQ (Kratos Analytical, Manchester, UK) MS instrument operated in positive reflectron mode, peptides were observed as [M+H]+ ions. For external calibration, bradykinin fragment 1–7, angiotensin II, ACTH fragment 18–39 and insulin oxidized beta chain, were used. Peptide mass fingerprinting. PeptideMass program was used to in silico cleave entered respective trastuzumab and rituximab light and heavy chain sequences with trypsin and to compute generated peptide masses with their specified modifications [8]. During database searching taxonomy parameter were not specified, selected enzyme was trypsin and only peptide masses higher than 500 Da have been displayed. Other search parameters were set

to allow up to 1 miss cleavage, methionines in oxidized form (variable modification), cysteines in carbamidomethylated form (fixed modification), average mass, peptide ions in protonated [M+H]+ form, mass tolerance (MT) parameter was set from ±1 to ±2 Da. Protein molecular mass was not specified. Enzymatic deglycosylation. Deglycosylation of rmAbs was performed according to the procedure described in Kamoda et al. [12]. Solutions of the pharmaceutical preparations, trastuzumab (Herceptin® ) 21 mg/ml and rituximab (MabThera® ) 10 mg/ml, collected from the vials immediately after clinical use in Vienna Hospital Rudolfstiftung, were dialyzed against MilliQ-water (3 days, 4 ◦ C) using cellulose dialysis tube (cut off 12,000–14,000 Da, Japan Medical Science, Tokyo, Japan). After dialysis, solutions were freeze dried and lyophilized and samples of pharmaceutical preparations, 0.5 mg each, were dissolved in 49 ␮l phosphate buffer (pH 7.5) to obtain a final concentration of 500 ␮g/50 ␮l. PNGase F of Flavobacterium meningosepticum, recombinant, from E. coli, 100 U/100 ␮l, (Roche Diagnostics GmbH, Mannheim, Germany) was used for the cleavage of N-linked oligosaccharides, (15 U/150 ␮g rmAb). Neuraminidase (sialidase) from Arthrobacter ureafaciens 1 U/100 ␮l, 30 mU/150 ␮l rmAb (Roche Diagnostics GmbH, Mannheim, Germany) was used for the cleavage of sialic acids and carboxypeptidase B, 750 U/ml, from pig pancreas (Roche Diagnostics GmbH, Mannheim, Germany) was used for the study of lysine truncation (0.75 U/150 ␮g rmAb). After digestion (overnight at 37 ◦ C) samples were stored at −80 ◦ C.

3. Results and discussion 3.1. 1D-SDS-PAGE analysis of trastuzumab and rituximab In order to examine relevant quality parameters, including detection, and purity assessment, 1-D SDS-PAGE experiments,

D. Nebija et al. / Journal of Pharmaceutical and Biomedical Analysis 56 (2011) 684–691

689

Table 1 Summary of MALDI-MS data for the identification of trastuzumab light and heavy chains. Rank of result

Database

Trastuzumab light chain 1 NCBInr

1

MSDB

2

MSDB

1

SwissProt

Trastuzumab heavy chain NCBInr 1

1

MSDB

Protein name/accession number Crystal structure of extracellular domain of human Her2 complexed with Herceptin Fab, chain A; gi28948772/1N8Z A Fab fragment of humanized antibody 4d5, version 4, chain A – human (1FVDA) Mixture of (a) peptide CH68665/CQ890830

(b) 1 FVCC(Fv fragment of humanized antibody, 4d5, version 8, chain C – human) Ig kappa chain C region, IGKC HUMAN

Mixture of (a) Enzyme deglycosylated human Igg1 Fc fragment, chain A; gi254574798/3DNK A

(b) Fab fragment of trastuzumab, gi442924/1FVE B Mixture of (a) Human CAC21795/CAC21795

Additional comments

protein sequence is identical with the constant domain of trastuzumab light chain (residues 108–214) variable fragment of trastuzumab light chain (residues 1–108) Sequence is identical with trastuzumab light chain residues 109–214

Amino acid sequence is identical with the Fc fragment of herceptin (residues 238–450); D-62 (aspartic acid) of this protein corresponds to trastuzumab asparagine residue, N-300 (non-deglycosylated form)

Protein sequence is identical to trastuzumab residues 218–450 (except C5S mutation) including CH2 and CH3 regions

(b) Fab fragment of trastuzumab, 1FVEB

under reducing and non-reducing conditions, were performed. Under reducing conditions both rmAbs were resolved in two distinct Mr species which migrated in two bands (upper band at ∼50 kDa and lower band ∼23 kDa) confirming the migration behavior typical for IgG antibodies which are composed of two identical subunits each composed by two polypeptide chains: two heavy and two light chains, linked via disulfide bonds (Fig. 3, gels A and B). Under non-reducing conditions, single bands of rmAbs at Mr ∼150 kDa have been observed (sum of two heavy and light chains) and in contrast to reducing conditions the band sharpness was reduced (Fig. 3, gels C and D). Sensitivity and linearity were assessed, as well. After N-deglycosylation, only a small decrease in Mr of rmAbs heavy chains was observed indicating the presence of N-linked glycans whereas Mr of light chains was not affected. For comparison purpose, deglycosylation experiments were performed for recombinant fusion protein abatacept (Fig. 3, gel E).

3.2. 2-DE analysis of trastuzumab and rituximab Recombinant mAbs are biological products derived from rDNA technology, produced in mammalian CHO cells. Therefore, it is expected to have a certain degree of heterogeneity and isoforms. As the heterogeneity of these products defines their quality, the characterization of the degree and profile of the heterogeneity constitutes an essential part of regulatory requirements [38].

Mass tolerance

Sequ. cov.

Score

Matched peptides

±1 Da

65%

219

18

±2 Da

73%

194

17

±2 Da ±2 Da

97%

166 106

18 10

±2 Da

65%

74

8

±1.2 Da ±1.2 Da

71%

226 136

28 16

55%

85

12

±1.2 Da ±1.2 Da

65%

222 135

26 16

±1.2 Da

55%

85

12

The applied 2-DE method was optimized for the pI gradient (IPG strips 3–11 NL, 6–9 L), gel concentration and visualization procedure. 2-DE gels of glycosylated antibodies revealed the characteristic glycoprotein migration behavior resulting in highly complex, poorly resolved spot patterns with small differences in Mr , but different pIs (Fig. 4). 2-DE analysis revealed that heavy chain spots of rmAbs migrated at Mr ∼50 kDa whereas their light chains migrated at Mr ∼23 kDa. Respective heavy chain pI ranges for trastuzumab and rituximab were 6.5–8.7 and 6.3–9.0; and respective light chain pI ranges were 5.7–8.1 and 6–8.5. Higher sensitivity of silver versus Coomassie staining and improved spot resolution at lower sample amounts was observed, as well (data not shown). No improvement in spot resolution could be documented applying IPG 6–9 L. No significant differences were observed in 2-DE spot patterns between N-deglycosylated, desialinated and carboxypeptidase B digested rmAbs and their glycosylated forms. Investigated rmAbs harbour only one glycosylation site, containing predominantly neutral N-linked glycans which contribute to only 2% in their Mr [39], in addition sialic acid content is only 1–5% of total glycan amount. Expected pI and Mr differences after N-linked glycan release are therefore too small to be unambiguously detected with 2-DE. Calculated pI differences between glycosylated and N-deglycosylated, and 450-lysine and truncated lysine variants is only ∼0.13 [8], which is also too small to be detected on 2-DE gels considering a certain variance regarding to gel-to-gel reproducibility. Respective

690

D. Nebija et al. / Journal of Pharmaceutical and Biomedical Analysis 56 (2011) 684–691

Table 2 Summary of MALDI-MS data for the identification of rituximab light and heavy chains. Rank of result

Database

Protein name/accession number

Additional comments

Mass tolerance

Sequ. cov.

Score

Matched peptides

Rituximab light chain 1

NCBInr

Fab of rituximab (crystal structure of Rituximab Fab in complex with an epitope peptide, chain L) gi 146387638/2OSL L

Human constant region (residues 107–213)

±1.1 ±1.5 ±1.7

35%42%45%

97141157

9

Rituximab heavy chain 1

NCBInr

Fc fragment of rituximab (bound to a minimized version of the B-Domain from protein A called Z34c, chain A); gi20664303/1L6X A

42% 57%

66 103

8

2-DE gels of N-deglycosylated trastuzumab and rituximab are presented in Fig. 4 (gels C and F, 2-DE gels of desialinated and carboxypeptidase B treated antibody samples are not shown). 3.3. MALDI-TOF MS and peptide mass fingerprinting analysis of rmAbs For identification purpose, tryptic in-gel digestion of selected silver stained 2-DE gel spots, corresponding to heavy and light chains of trastuzumab (Fig. 4, gel C) and rituximab (Fig. 4, gel F), respectively, was performed. Peptide masses determined by MALDI-TOF MS analysis were compared to in silico (theoretical) peptide masses by applying the enzyme cleavage rules, from three amino acid sequence databases: MSDB, NCBInr and SwissProt using MASCOT [40] and PROWL (ProFound) [41] search engines. Antibody concentration, amount of trypsin, digestion time, extraction of generated peptides and other factors related to the MALDI-TOF sample preparation have been studied and optimized. In Figs. 5 and 6, respective MALDI-TOF spectra of generated peptides from deglycosylated light and heavy chains of trastuzumab and rituximab are presented. 3.3.1. MALDI-TOF MS and peptide mass fingerprinting analysis of trastuzumab For the identification of trastuzumab light chain, a list of 18 peptide masses determined by MALDI-TOF MS from corresponding protein spots were compared to data from sequence databases NCBInr, SwissProt and MSDB using the search engines MASCOT and PROWL [40,41] (Fig. 4C). The generated top score results for trastuzumab light and heavy chain including important MS parameters are shown in Table 1. For the identification of trastuzumab heavy chain, a list of 29 peptide masses determined by MALDI-TOF MS was compared to the same databases using MASCOT and PROWL/ProFound [40,41] search engines (Table 1). 3.3.2. MALDI-TOF MS and peptide mass fingerprinting analysis of rituximab For amino acid sequence confirmation of rituximab light chain, a list of 13 peptide masses experimentally determined by MALDITOF MS were compared to sequence databases as described above. Results are summarized in Table 2. The databases MSDB and SwissProt, respectively, generated the highest score results for IgG kappa light chain and fragment of monoclonal antibodies suggesting that in this databases rituximab light chain sequence information is not mentioned. For amino acid sequence confirmation of rituximab heavy chain, a list of 14 peptide masses experimentally determined by MALDITOF MS was compared to the same databases (Table 2). SwissProt and MSDB generated similar results.

±1.5

±2.0

12

13

11

Some MALDI spectra are available as supporting information. 4. Concluding remarks Experimental results show that 1-D SDS-PAGE analysis can be used as a simple, fast and reliable method for the quality assessment of rmAbs, trastuzumab and rituximab. This technique allows the estimation of antibody Mr and purity. In contrast to 1-D SDSPAGE, 2-DE analysis delivered more comprehensive information regarding identity and structural integrity of rmAbs, and apart from the fast estimation of Mr and pI, 2-DE proved to be a very suitable technique for the illustration of charge heterogeneity, as well. Experimental data revealed that MALDI-TOF peptide mass fingerprinting could be used as rapid method for the identification of rmAb characteristics, provided that the sequence information is present in respective databases. Apart from its application during the production process of rmAbs, 2-DE in combination with subsequent MALDI-TOF MS analysis in our opinion might be used as a routine technique in quality control laboratories, suited in the regulatory/pharmacopoeia environment, as well. Acknowledgements D.N. thanks the Office of Research Services and International Relations of University of Vienna for the financial support. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jpba.2011.07.006. References [1] D.J. Slamon, W. Godolphin, L.A. Jones, J.A. Holt, S.G. Wong, D.E. Keith, W.J. Levin, S.G. Stuart, J. Udove, A. Ullrich, M.F. Press, Studies of the HER-2/neu protooncogene in human breast and ovarian cancer , Science 244 (1989) 707–712. [2] C.A. Hudis, Trastuzumab—mechanism of action and use in clinical practice , N. Engl. J. Med. 357 (2007) 39–51. [3] R.J. Harris, B. Kabakoff, F.D. Macchi, F.J. Shen, M. Kwong, J.D. Andya, S.J. Shire, N. Bjork, K. Totpal, A.B. Chen, Identification of multiple sources of charge heterogeneity in a recombinant antibody , J. Chromatogr. B 752 (2001) 233–245. [4] P.J. Carter, L.G. Presta, Method for making humanized antibodies. United States Patent, October 28, 2003. US patent no. 6,639,055, U.S. Patent and Trademark Office, Washington, DC. http://patft.uspto.gov/ (accessed: 18.08.10). [5] M.E. Reff, K. Carner, K.S. Chambers, P.C. Chinn, J.E. Leonard, R. Raab, R.A. Newman, N. Hanna, D.R. Anderson, Depletion of B cells in vivo by a chimeric mouse human monoclonal antibody , Blood 83 (1994) 435–445. [6] J. Du, H. Wang, C. Zhong, B. Peng, M. Zhang, B. Li, S. Hou, Y. Guo, J. Ding, Crystal structure of chimeric antibody C2H7 Fab in complex with a CD20 peptide , Mol. Immunol. 45 (2008) 2861–2868. [7] www.ema.europa.eu/htms/human/epar/a.htm (accessed 10.09.10). [8] M.R. Wilkins, I. Lindskog, E. Gasteiger, A. Bairoch, J.C. Sanchez, D.F. Hochstrasser, R.D. Appel, Detailed peptide characterisation using peptide mass—a WorldWide Web accessible tool , Electrophoresis 18 (1997) 403–408.

D. Nebija et al. / Journal of Pharmaceutical and Biomedical Analysis 56 (2011) 684–691 [9] H.Z. Wan, S. Kaneshiro, J. Frenz, J. Cacia, Rapid method for monitoring galactosylation levels during recombinant antibody production by electrospray mass spectrometry with selective-ion monitoring , J. Chromatogr. A 913 (2001) 437–446. [10] J. Hodoniczky, Y.Z. Zheng, D.J. James, Control of recombinant monoclonal antibody effector functions by Fc N-glycan remodeling in vitro , Biotechnol. Prog. 21 (2005) 1644–1652. [11] M. Noze, H. Wigzell, Biological significance of carbohydrate chains on monoclonal antibodies , Proc. Natl. Acad. Sci. U.S.A. 80 (21) (1983) 6632–6636. [12] S. Kamoda, C. Nomura, M. Kinoshita, S. Nishiura, R. Ishikawa, K. Kakehi, N. Kawasaki, T. Hayakawa, Profiling analysis of oligosaccharides in antibody pharmaceuticals , J. Chromatogr. A 1050 (2004) 211–216. [13] S. Kamoda, R. Ishikawa, K. Kakehi, Capillary electrophoresis with laser-induced fluorescence detection for detailed studies on N-linked oligosaccharide profile of therapeutic recombinant monoclonal antibodies , J. Chromatogr. A 1133 (2006) 332–339. [14] L.A. Gennaro, O. Salas-Solano, On-line CE-LIF-MS technology for the direct characterization of N-linked glycans from therapeutic antibodies , Anal. Chem. 80 (10) (2008) 3838–3845. [15] S. Ma, W. Nashabeh, Carbohydrate analysis of a chimeric recombinant monoclonal antibody by capillary electrophoresis with laser-induced fluorescence detection , Anal. Chem. 71 (22) (1999) 5185–5192. [16] M. Nakano, D. Higo, E. Arai, T. Nakagawa, K. Kakehi, N. Taniguchi, A. Kondo, Capillary electrophoresis-electrospray ionization mass spectrometry for rapid and sensitive N-glycan analysis of glycoproteins as 9-fluorenylmethyl derivatives , Glycobiology 19 (2) (2009) 135–143. [17] J. Stadlmann, M. Pabst, D. Kolarich, R. Kunert, F. Altmann, Analysis of immunoglobulin glycosylation by LC-ESI-MS of glycopeptides and oligosaccharides , Proteomics 8 (2008) 2858–2871. [18] J.A. Kelly, C.S. Lee, On-line post-capillary affinity detection of immunoglobulin G subclasses and monoclonal antibody variants for capillary electrophoresis , J. Chromatogr. A 790 (1997) 207–214. [19] K.G. Moorhouse, W. Nashabeh, J. Deveney, N.S. Bjork, M.G. Mulkerrin, T. Ryscamp, Validation of an HPLC method for the analysis of the charge heterogeneity of the recombinant monoclonal antibody IDEC-C2B8 after papin digestion , J. Pharm. Biomed. Anal. 16 (1997) 593–603. [20] P.V. Beum, A.D. Kennedy, R.P. Taylor, Three new assays for rituximab based on its immunological activity or antigenic properties: analyses of sera and plasmas of RTX-treated patients with chronic lymphocytic leukemia and other B cell lymphomas , J. Immunol. Methods 289 (2004) 97–109. [21] M.A. Bookman, K.M. Darcy, D.C. Pearson, R.A. Boothby, I.R. Horowitz, Evaluation of monoclonal humanized anti-HER2 antibody, trastuzumab, in patients with recurrent or refractory ovarian or primary peritoneal carcinoma with overexpression of HER2: a Phase II Trial of the Gynecologic Oncology Group , J. Clin. Oncol. 21 (2003) 283–290. [22] H.J. Stemmler, M. Schmitt, A. Willems, H. Bernhard, N. Harbeck, V. Heinemann, Ratio of trastuzumab levels in serum and cerebrospinal fluid is altered in HER2positive breast cancer patients with brain metastases and impairment of bloodbrain barrier , Anticancer Drugs 18 (2007) 23–28. [23] http://www.accessdata.fda.gov/scripts/cder/drugsatfda/ (accessed 10.09.10). [24] P. Carter, L. Presta, C.M. Gorman, J.B.B. Ridgway, D. Henner, W.L.T. Wong, A.M. Rowland, C. Kotts, M.E. Carver, H.M. Shepard, Humanization of an anti-p185HER2

[25]

[26]

[27]

[28]

[29]

[30] [31]

[32]

[33] [34] [35] [36]

[37]

[38]

[39]

[40]

[41]

691

antibody for human cancer therapy , Proc. Natl. Acad. Sci. U.S.A. 89 (1992) 233–245. C.W.N. Damen, H. Rosing, J.H.M. Schellens, J.H. Beijnen, Quantitative aspects of the analysis of the monoclonal antibody using high-performance liquid chromatography coupled with electrospray mass spectrometry , J. Pharm. Biomed. Anal. 46 (2008) 449–455. G. Hunt, K.G. Moorhouse, A.B. Chen, Capillary isoelectric focusing and sodium dodecyle sulfate capillary gel electrophoresis of recombinant humanized monoclonal; antibody HER2 , J. Chromatogr. A 744 (1996) 295–301. R.J. Harris, Processing of C-terminale lysine and arginine residues of proteins isolated from mammalian cell culture , J. Chromatogr. A 705 (1995) 129–134. X.M. Lam, J.Y. Jang, J.L. Cleland, Antioxidants for prevention of metionone oxidation in recombinant monoclonal antoibody HER2 , J. Pharm. Sci. 6 (1997) 1250–1255. R.J. Harris, A.A. Murnane, S.L. Utter, K.L. Wagner, E.T. Cox, G.D. Polastry, J.C. Helder, M.B. Sliwkowski, Assessing genetic heterogeneity in production of cell line: detection by peptide mapping of a low level Tyr to Gln Sequence variant in a recombinant antibody , Nat. Biotechnol. 11 (1993) 1293–1297. W.J. Henzel, C. Watanabe, J.T. Stults, Protein identification: origins of peptide mass fingerprinting , J. Am. Soc. Mass Spectrom. 14 (2003) 931–942. W. Schlags, B. Lachmann, M. Walther, M. Kratzel, C.R. Noe, Two-dimensional electrophoresis of recombinant human erythropoietin: a future method for the European Pharmacopoeia? Proteomics 2 (2002) 679–682. T. Rabilloud, C. Valette, J.J. Lawrence, Sample application by in-gel rehydration improves the resolution of two-dimensional electrophoresis with immobilized pH gradients in the first dimension , Electrophoresis 15 (1994) 1552–1558. A. Görg, W. Postel, S. Günther, The current state of two-dimensional electrophoresis with immobilized pH gradients , Electrophoresis 9 (1988) 531–546. U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage , Nature 227 (1970) 680–685. H. Blum, H. Beier, H.J. Gross, Improved silver staining of plant-proteins, RNA and DNA in polyacrylamide gels , Electrophoresis 8 (1987) 93–99. A. Shevchenko, M. Wilm, O. Vorm, M. Mann, Mass spectrometric sequencing of proteins from silver-stained polyacrylamide gels , Anal. Chem. 68 (5) (1996) 850–858. F. Garahdaghi, C.R. Weinberg, D.A. Meagher, B.S. Imai, S.M. Mische, Mass spectrometric identification of proteins from silver-stained polyacrylamide gel: A method for the removal of silver ions to enhance sensitivity , Electrophoresis 20 (1999) 601–605. International Conference on Harmonization. Harmonized tripartite guideline. specifications: test procedures and acceptance criteria for biotechnological/biological products, Q6B. Step 5, Version March 1999. http://www.ich.org/LOB/media/MEDIA432.pdf (accessed 18.08.10). T.S. Raju, J.B. Briggs, S.M. Borge, A.S. Jones, Species-specific variation in glycosylation of IgG: evidence for the species-specific sialylation and branch-specific galactosylation and importance for engineering recombinant glycoprotein therapeutics , Glycobiology 10 (2000) 477–486. D.N. Perkins, D.J.C. Pappin, D.M. Creasy, J.S. Cottrell, Probability-based protein identification by searching sequence databases using mass spectrometry data , Electrophoresis 20 (1999) 3551–3567. http://prowl.rockefeller.edu/prowl-cgi/profound.exe (accessed 18.08.10).